[Fall risk assessment in regular exercising elderly women].

Aim: Fall prevention is important for elderly people to maintain their functional independence. We made a longitudinal fall-risk assessment using our "Fall-predicting score" of women who are 60 years or older and who exercised regularly.

Methods: We sent "fall-predicting questionnaires" to 632 elderly women aged 60 years or older (mean 65.0+/-4.3), members of "Miishima gymnastics program", and asked about their fall history of falling in the past year in 2004 and 2005. We performed a logistic regression analysis to determine the future risk factor of falling in 2005.

Results: The number of people who fell was 134 (21.2%) in 2004 and 121 (19.1%) in 2005. The number of people who fell decreased in the seventh decade, but increased in the eighth decade, and members for 6-10 years showed most decreased fall rates. Logistic regression analysis revealed that age, falls in 2004, "tripping", "cannot squeeze a towel", and "walk steep slope around the house" were significant independent risk factors of "falls in 2005". Logistic regression analysis of non-fallers in 2004 showed that age and "tripping" were the significant independent risk factors of "falls in 2005", and the analysis of people who fell in 2004 showed that age, "tripping", "cannot squeeze a towel", "walk steep slope around the house", and "taking more than 5 medicines" were significant independent risk factors for falls in 2005.

Conclusions: In regular exercising elderly women, exercise appears to prevent falls in people in the seventh decade and in the members of 6-10 years. Age, past history of falls, and fall-predicting questionnaire were important risk predictors of future falls.